Cancer is the second leading cause of death in the United States, and while most current cancer therapies target the primary tumor, it is estimated that 90% of cancer-associated deaths are due to the process of metastasis. Throughout this process, tumor cells alter their presentation of surface molecules in a manner that permits inappropriate interactions with their microenvironment and promotes invasion and systemic dissemination. Concurrently, soluble factors released from the tumors act in paracrine and endocrine fashions to mobilize stromal populations, consisting largely of pro-tumorigenic myeloid cells, to the blood and distant sites of disease. These populations promote the establishment of metastases in those sites through conditioning of the microenvironment and secretion of tolerogenic cytokines to prevent anti-tumor immune responses. One particular cell- surface alteration that occurs on metastasizing tumor cells is the increased presentation of a carbohydrate motif known as the Thomsen-Friedenreich Antigen (T-Antigen). We hypothesize that metastatic cells interact with stromal populations in the metastatic niche by binding of the T Antigen to galectin-3 on the surfaces of these stromal cells. Thus, goals of this proposal are twofold: (1) to determine how binding of the T-Antigen to galectin-3 promotes interactions with pro- tumorigenic immune cells in the metastatic niche thus inducing tolerance and (2) to develop a metastasis vaccine that specifically targets the T-Antigen to facilitate the production of anti-tumor immune responses with high specificity. The development of cancer vaccines represents significant challenges over that of conventional vaccines in that there exists a limited repertoire of non-self-antigens that form suitable targets for anti-tumor immune responses. Furthermore, immunoediting through the selection of tumor cells that have downregulated, or do not express, the target antigen represents an effective mechanism of escape for the tumors. The ubiquitous representation of the T-Antigen on the majority of carcinomas, combined with its lack of expression on normal cells, make it an ideal vaccine target. Furthermore, that its presentation is the product of complex networks of glycosyltransferases, rather than one individual gene, renders immunoediting by tumor cells extremely challenging. In this work, we will generate glycopeptide-based vaccines and use these therapeutics in both prophylactic and treatment settings in a genetic mouse model of lung adenocarcinoma. We expect that this work will facilitate the generation of a new class of cancer therapeutics that specifically target changes in glycosylation in advanced stage malignancies.